Purpose in life promotes resilience to age-related brain burden in middle-aged adults

Data was obtained from 624 middle-aged adults (mean age 53.71 ± 6.9 years; age range: 42–67; 303 women; mean years of education [YoE] 17.1 ± 3.8 years, YoE range: 8–34) from the Barcelona Brain Health Initiative (BBHI; https://​bbhi.​cat/​en/​). This is an ongoing longitudinal cohort study investigating the determinants of brain and mental health in middle-aged individuals [3, 49]. For the present work, participants were included if they met the following criteria: (i) completed a PiL questionnaire (see below) and a neuropsychological assessment and (ii) T1- and T2-weighted and resting-state functional magnetic resonance imaging (rs-fMRI) scans were available. Participants were excluded if they (i) had any neurological or psychiatric diagnosis, (ii) had below normative data on any of the administered neuropsychological tests, (iii) imaging quality check was not satisfactory, and (iv) had statistically extreme values on WMLs, based on the Shapiro-Wilk test (p > 0.05 [50, 51]) and a visual inspection of the box plots.

PiL was measured through an online self-administered questionnaire using Ryff’s Psychological Well-Being Scale [20]. Within the BBHI framework, this scale was also used to assess personal growth. Additionally, access to further well-being measures was available (please, see [3]). However, in the present study, we only focused on the PiL dimension, as previous investigations have particularly revealed its potential to confer resilience to brain burden (i.e., [34]). In the PiL questionnaire, participants reported from 1 (strongly disagree) to 5 (strongly agree) the following questions: “I live life one day at a time and don’t really think about the future”; “I have a sense of direction and purpose in life”; “I don’t have a good sense of what it is I’m trying to accomplish in life”; “My daily activities often seem trivial and unimportant to me”; “I enjoy making plans for the future and working to make them a reality”; “Some people wander aimlessly through life, but I am not one of them”; and “I sometimes feel as if I’ve done all there is to do in life.” Direct and inverse items were controlled to obtain a total sum score per participant. All included participants answered the complete number of questionnaire items.

Neuropsychological testing was administered by expert neuropsychologists in a single session of approximately 90 min [3, 17]. Tests battery followed a fixed order and included direct and inverse Digit Span [52], Trail Making Test parts A and B [52], Reasoning Matrix [53], Rey Auditory-Verbal Learning Test [54], Block Design Test [53], Letter-Number Sequencing [52], Digit-Symbol Substitution Test and Cancellation subtests from WAIS-IV [53], and Corsi block-tapping test [52].

MRI data were acquired in a 3-T Siemens scanner (MAGNETOM Prisma) with a 32-channel head coil at Unitat d’Imatge per Ressonància Magnètica IDIBAPS (Institut d’Investigacions Biomèdiques August Pi i Sunyer) at Hospital Clínic de Barcelona, Barcelona. For all participants, a high-resolution T1-weighted structural image was obtained with a magnetization-prepared rapid acquisition gradient echo (MPRAGE) three-dimensional protocol (repetition time [TR] = 2400 ms, echo time [TE] = 2.22 ms, inversion time = 1000 ms, field of view [FOV] = 256 mm, flip angle = 8° and 0.8-mm isotropic voxel). Additionally, a high-resolution 3D SPC T2-weighted structural brain MRI was undertaken (TR = 3200 ms, TE = 563 ms, flip angle = 120°, 0.8-mm isotropic voxel, FOV = 256 mm). They also underwent rs-fMRI multiband (anterior-posterior phase-encoding; acceleration factor = 8) interleaved acquisitions (T2-weighted EPI scans, TR = 800 ms, TE = 37 ms, 750 volumes, 72 slices, slice thickness = 2 mm, FOV = 208 mm). All MRI images were examined by a senior neuroradiologist for any clinically significant pathology and visually inspected by trained MRI technicians for subjective quality control of metallic or motion artifacts. To control for movement between rs-fMRI scans, the framewise displacement (FWD) mean was computed.

The FMRIB Software Library (FSL, version 5.0.11; https://​fsl.​fmrib.​ox.​ac.​uk/​fsl/​fslwiki/​), Statistical Parametric Mapping (SPM, version 12; https://​www.​fil.​ion.​ucl.​ac.​uk/​spm/​), and FreeSurfer (version 6.0; https://​surfer.​nmr.​mgh.​harvard.​edu/​) were used for preprocessing and analyzing MRI data. Preprocessing pipeline and head movement considerations are described in SM.

Data analyses were performed using IBM SPSS (IBM Corp. Released 2020. IBM SPSS Statistics, version 27.0. Armonk, NY: IBM Corp) and GraphPad Prism (version 9.0.0, GraphPad Software, San Diego, CA, USA).

First, the total sample was stratified to create two extreme groups, in a similar manner as previous PiL investigations (i.e., [23, 32‐34]). With this aim, we used the “visual binning” function from the SPSS, which divides all the included subjects according to a specified number of cut points. The option “equal percentiles based on scanned cases” was used to create 5 sub-groups: Q1 (PiL values: ≤ 21; N = 146), Q2 (PiL values: 22–25, N = 135), Q3 (PiL values: 26–27, N = 114), Q4 (PiL values: 28–30, N = 129), and Q5 (PiL values: ≥ 31, N = 100). We later conducted analyses focusing on the extreme groups: the Q1, named lower PiL (LP) group, and the Q5, named higher PiL (HP) group. These two groups shaped our sample of interest, conformed by 246 individuals. Basic demographic data (age, gender, YoE) was directly compared between the PiL groups through a one-way analysis of variance (ANOVA) and a chi-squared test. Cognitive data was integrated into three composite scores: an episodic memory composite (EMc), an executive functioning composite (EFc), and a working memory composite (WMc). The EMc was calculated considering the three recall measures from the Rey Auditory-Verbal Learning Test (immediate, delayed, and recognition). The EFc measure was computed considering the Reasoning Matrix, Block Design Test, Digit-Symbol Substitution Test and Cancellation subtests from WAIS-IV. The WMc measure was calculated with the inverse Digit Span and the Letter-Number Sequencing. Composites were obtained through factorial analyses with SPSS. Brain burden was evaluated considering the total estimation of WMLs. rs-fMRI analyses were computed using SyS data on the 14 Shirer circuits [59]. To investigate the neuropsychological differences between the PiL groups, a multivariate general linear model (GLM) was conducted considering all cognitive measures together as dependent variables. Moreover, a univariate GLM with WMLs as the dependent variable was calculated to study the brain burden group differences. Subsequently, Pearson correlations between cognitive status and WMLs in each group, as well as slope differences between the groups, were computed. This latter analysis was conducted with regression functions from GraphPad Prism. In addition, to investigate rs-fMRI differences, a multivariate GLM was undertaken considering the 14 Shirer networks altogether. Whether significant group differences emerged on a whole functional system, a subordinate zoom-in was conducted focused on its ROI-to-ROI functional couplings through multiple GLM analyses, considering within- and between-brain network connectivity. As per its exploratory nature, these analyses were not corrected for multiple comparisons. Finally, Pearson correlations were calculated to relate rs-fMRI measures with cognitive performance in each group. In all the stated statistical analyses, age and gender were used as covariates. YoE were included as a covariate when cognitive data was examined. Moreover, all rs-fMRI explorations were also controlled for FWD. All statistical analyses were two-tailed, and α was set at 0.05. For the Pearson correlation analyses, a bootstrapping with 5000 samples was also applied, and the bias-corrected and accelerated 95% confidence interval (CI) was reported. Quality checks were conducted using the stated covariates to corroborate whether the main associations were also present in the whole sample, as well as to further investigate the role of specific variables (i.e., age). Data in plots are presented with standardized Z scores, considering the main variables (cognition, WMLs, rs-fMRI) as well as the covariates included in each model (age, gender, YoE, FWD).

Neuropsychological status and WMLs did not differ between the HP and LP groups (all p-values > 0.05). However, the slopes between the groups (HP vs. LP) were significantly different in the association between EFc and WMLs (F = 9.957, p = 0.002). Greater WMLs were negatively associated with EFc in the LP group (r = − 0.283, p < 0.001, 95% CI [− 0.405, − 0.156]) but not in the HP group (r = 0.120, p = 0.243, 95% CI [− 0.093, 0.324]; Fig. 1). No significant results were observed for the other cognitive measures (all p-values > 0.05). As a quality check, in the whole sample, the negative association between EFc and WMLs was also present (r = − 0.137, p < 0.001, 95% CI [− 0.211, − 0.060]). Moreover, age, which did not differ between the groups (p > 0.05; Additional file 1: Table S1), was positively associated with WMLs in both groups (HP group: r = 0.345, p < 0.001, 95% CI [0.169, 0.499]; LP group: r = 0.384, p < 0.001, 95% CI [0.223, 0.525]). Demographic data (age, gender, YoE) in each PiL group is further displayed in Additional file 1: Table S1.

Subjects in the HP group showed lower SyS on the dorsal default-mode network than LP subjects (F = 4.907, p = 0.028), which indicated lesser segregation of this network from other brain circuits. Specifically, HP individuals had greater inter-network connectivity between dDMN nodes and the rest of the brain on the functional couplings depicted in Fig. 2A (see also Additional file 1: Table S2). It is worth noting that, out of the total functional connections identified in this subsequent analysis, three nodes were central: the right superior frontal node, which involved 15 out of 45 of the couplings (33.3%); the hippocampal formation, implicated in 14 out of 45 connections (31.1%), particularly the right hippocampus; and the midcingulate cortex, present in 8 out of 45 functional bridges (17.8%). Altogether, these functional hubs were present in 37 of the 45 functional connections, explaining 82.2% of the results. Conversely, subjects in the LP group exhibited more connectivity than HP individuals across DMN-like regions, particularly those involving connections between the dDMN and the ventral DMN (vDMN; Fig. 2B; Additional file 1: Table S2). The stated DMN-focused ROI-to-ROI analyses comprised a total of 765 comparisons. Of note, no differences were detected in the dDMN between-network connectivity with the other brain circuits nor in the connectivity within the dDMN nodes. Hence, our SyS results were driven by specific dDMN hubs with enhanced inter-connectivity with other brain regions. As a quality check, the negative association between PiL and dDMN SyS was also detected in the whole sample (r = − 0.081, p = 0.044, 95% CI [− 0.158, − 0.001]).

Finally, we investigated whether rs-fMRI might differently subtend cognitive function for a given brain burden estimation as a function of the PiL group. Considering the HP > LP identified couplings, we detected two functional connections positively associated with executive performance in the HP group. Both connections involved the midcingulate region of the dDMN and two other cognitive brain systems, the posterior salience (PS), within its thalamic node (r = 0.205; p = 0.045, 95% CI [0.021, 0.390]; Fig. 3A), and the precuneus network, within its midcingulate and posterior cingulate cortices (r = 0.206, p = 0.044, 95% CI [0.019, 0.374]; Fig. 3B). No associations with cognition were observed in the LP group (all p-values > 0.05). These correlations survived when controlling for WMLs (r = 0.204, p = 0.047, 95% CI [0.015, 0.378]; r = 0.214, p = 0.037, 95% CI [0.042, 0.373], respectively). Furthermore, greater values on the variables used in this analysis (EFc and rs-fMRI) were not directly related to WMLs (all p-values > 0.05).

The main result of this investigation was that the association between brain burden and executive performance depends on PiL estimates in middle-aged adults, with subjects with higher PiL exhibiting greater resilience to WMLs. No associations were found regarding episodic or working memory. This aligns with previous literature revealing that WMLs are especially related with executive function and processing speed domains (for a review, see [60]). Furthermore, regarding PiL, previous studies have examined the relationships between this construct and cognitive status, mainly in older samples, with comparable results (i.e., [16, 30, 32]). In a large epidemiologic study of aging, greater sense of purpose was associated with slower rates of cognitive decline and reduced risk of mild cognitive impairment (MCI) and AD [32]. A recent meta-analysis has reinforced these assumptions, revealing that PiL is associated with a reduced risk of dementia [31]. Boyle et al. [34] observed, at the brain level, that superior levels of PiL reduced the deleterious effects of AD pathologic changes on cognitive function. These results suggest that PiL might be a relevant resilience factor, allowing it to counteract the deleterious impact of brain aging and disease on cognitive functions [5, 6]. To the best of our knowledge, this is the first study showing that the PiL-dependent association between brain burden and cognition is also present in healthy middle-aged individuals. Therefore, since according to the current operational definitions (https://​reserveandresili​ence.​com/​), PiL influenced the associations between a measure of brain burden and cognitive status, it can be considered as a key CR factor, reinforcing the notion that resilience can be influenced by multiple genetic and environmental factors operating continuously across the lifespan [5, 6].

The neurobiological mechanisms explaining the pathways by which PiL might promote resilience are poorly understood. Different explanatory models have been proposed to elucidate how PiL relates to cognitive health (i.e., [61]). It is possible that superior PiL estimates are linked to healthier behaviors [27, 28], which are in turn related to better cognitive performance and lower odds of dementia [62]. Alternatively, PiL might relate to biological mechanisms that influence cognitive performance [63]. Thus, scoring higher in PiL has been reported to be associated with lower interleukin-6 (IL-6) plasma concentration levels [64]. Moreover, in middle-aged [65] and older adults [66], greater PiL was found to be related to lower levels of hemoglobin A1c (HbA1c). Also, Ryff’s Psychological Well-Being Scale measures (including PiL) were examined regarding distinct physiological processes (i.e., blood pressure, urinary catecholamines, and salivary cortisol) in a young and middle-aged sample. In this investigation, data revealed that greater PiL is associated with lower total cortisol levels [67]. This is relevant as inflammation, glucose, and cortisol regulation are associated to cognitive and brain health during the lifespan [68‐70].

Focusing on brain measures, previous observations have revealed that PiL is linked to a lower risk of cerebrovascular conditions [71, 72], which might confer a resistance role to PiL (i.e., [73]). Furthermore, PiL might help cope with neural damage, providing resilience to brain burden (i.e., [32, 34]). This proposal is consistent with our data, as WMLs were linked to the cognitive status contingent on PiL estimates. Within this context, our study went further on the investigation of the functional brain mechanisms through which PiL may promote cognition, pointing to a particular emphasis on DMN. This aligns with previous investigations that have explored the functional architecture of purpose and sense of life meaning and have revealed a central role of DMN, an archetypical network, which is critically involved in autobiographical remembering, self-referential thought, mental simulation, and mind wandering (i.e., [74‐77]). In this sense, Mwilambwe-Tshilobo et al. [78] found that high levels of meaning in life correlated with increased, and more modular, connectivity between the DMN and the limbic system. Additionally, we also identified key regions within the dDMN associated with superior levels of PiL, even though these topographic results should be further clarified. One of them was the hippocampus, a complex brain structure embedded deep into the temporal lobe. This result links with the observations of Waytz et al. [79], reporting that meaning in life is associated to greater connectivity of the medial temporal lobe subsystem of the DMN.

Overall, while previous findings reinforce the associations between PiL and functional brain measures associated to cognitive function, in our study, we did not found evidence that differences in functional connectivity between the PiL groups, or the associations between HP-related rs-fMRI characteristics and cognitive status, were associated to WMLs burden. These findings therefore indicate that while greater dDMN connectivity among subjects with superior PiL conferred a cognitive advantage on executive functions, this was independent of the deleterious effect of WML in this cognitive domain, and that the brain network’s integrity status was not directly explaining the attenuating effect of high PiL in the relation between WML burden and cognition. Note that these findings differ from those of Ewers et al. [43], where both among familial and sporadic AD, measures of SyS were found to attenuate the effect of pathology (estimated by years to symptom onset in the former and measured by tau-PET in the second) on cognitive function. Within this context, and while our main findings do not challenge the fact that PiL confers resilience to the deleterious effect of WML in middle age, they also suggest that the specific functional mechanism identified here may reflect a brain reserve mechanism (see https://​reserveandresili​ence.​com/​) but may not involve an active adaptation of functional cognitive processes in the presence of the brain burden measure investigated in the present report.

While the results of the present study highlight that the brain mechanisms through which high PiL confers resilience in middle age need to be investigated in further research, a relevant aspect is that PiL is a potentially modifiable psychological factor. In this context, psychological interventions, such as meaning-centered psychotherapy (MCP), an extension of classic Frankl’s logotherapy further informed by the contributions of Yalom [80], may be implemented to enhance meaning, spiritual well-being, and quality of life [81‐84]. Acceptance and commitment therapy (ACT) might also help individuals to live meaningful lives by encouraging their engagement in activities that are consistent with their values (i.e., [48]). It is relevant to note that improvements in mental health conditions (i.e., anxiety disorders) from psychological therapy have been shown to be associated with reduced incidence of future dementia [85]. Hence, and as the neurobiological underpinnings of both protective and risk psychological factors (i.e., [19, 86]) are being unveiled, such interventional approaches might help to understand how psychological therapies may promote a healthy brain during the lifespan and aid in the prevention of cognitive impairment later in life.

The main limitation of the present study was its cross-sectional nature, which did not allow to explore whether PiL may operate as a resilience mechanism on age-related cognitive decline further than in immediate cognitive status. This also constrains our capacity to infer directionality on the explored variables. Moreover, it is worth highlighting that the use of extreme groups in the present study, as done in previous investigations in the field (i.e., [34]), while allowing the study of PiL concept in a thorough manner, also implies a relevant loss of sample. Finally, part of the neuroimaging results revealing the neurobiological underpinnings of PiL was obtained in an exploratory manner. In particular, the associations between rs-fMRI connectivity measures and cognitive status suggesting a brain reserve effect among PiL needs to be further investigated in forthcoming studies, and in distinct populations (i.e., in the AD continuum).